THE MINISTRY OF EDUCATION AND TRAINING THE MINISTRY OF NATIONAL DEFENSE
VIETNAM MILITARY MEDICAL UNIVERSITY
THAI VIET TANG
THE STUDY ON BONE MINERAL DENSITY AND RISK FACTORS OF FRACTURE IN POSTMENOPAUSAL WOMEN IN RACH GIA CITY, KIEN GIANG PROVINCE
Specialist: Internal Medicine Code: 9720107
SYNOPSIS OF DOCTORAL DISSERTATION
HA NOI – 2019
The work has been successfully completed at Vietnam Military Medical University
Science Instructors: Assoc. Prof., Ph.D. Đoàn Văn Đệ
Opponent 1: Opponent 2: Opponent 3:
The thesis has been defended at Institutelevel Thesis Evaluation Council at Military Medical University............ (hour),...../...../..... (date)
This thesis may be found at: Vietnamese National Library Library of Military Medical University
1
INTRODUCTION
THE URGENT NATURE OF THE THESIS
Osteoporosis is defined as a pathology characterized by reduced
bone strength and an increased risk of bone fractures. Bone strength
is related to two main factors which are bone mineral density (BMD)
and bone structure. In postmenopausal women and men over 50,
BMD is reduced by age, and the structure of the bone is degraded.
The reduction of BMD and bone structure degradation make the bone
weak and easily broken when impacted with a small force (such as
sneezing). Therefore, fractures are a consequence of osteoporosis.
Osteoporosis bone fracture is a major medical problem in the
elderly. Worldwide, there are more than 8.9 million people with bone
fracture every year; in which women are the majority (61%). Bone
fractures, especially femoral fracture, increase the risk of mortality.
The patients after bone fractures have a poor quality of life and are
unable to walk normally. Bone fracture is also a global economic
burden, which the annual cost associated with treatment in the United
States is up to 1020 billion USD, 2.7 billion EUR in the UK and 7.5
USD in Australia. Patients with bone fracture, especially the femoral
neck fracture suffer from complications such as pain, disability and
1220% of mortality in the first year. The survivors are also greatly
reduced the quality of life.
Currently, there are many methods of diagnosing osteoporosis,
in which bone density measurement with DXA method has
considered as a gold standard. Individuals whose BMD decline more
than 2.5 in standard deviations in comparison with the average value
2
at the age of 2030 are diagnosed with osteoporosis. Patients
diagnosed with osteoporosis are indicated for treatment. In addition,
patients with a history of bone fractures who have not had
osteoporosis are also indicated for treatment.
But osteoporosis only partially explains the total number of
fractures. Indeed, 55% of women suffering a fracture and 70% of
men suffering from fracture, but they do not have osteoporosis.
Therefore, osteoporosis only explains about 45% in women and 30%
in men having fractures. Many studies around the world have shown
that in addition to osteoporosis (or reduction of BMD), other factors
also associated with fractures: old age, women, smoking, excessive
alcohol use, weight loss., reduced height, history of fractures, long
term corticosteroid use, rheumatoid arthritis, secondary osteoporosis,
and falls. Therefore, besides BMD, there are 12 other factors that can
help assess an individual's risk of bone fracture.
For 10 years, there have been some prognostic models
developed to assess the risk of fracture. Two popular models are
Garvan Fracture Risk Calculator (Garvan) and FRAX. Garvan model
uses 5 risk factors (age, BMD, weight loss, history of fractures, and
falls); FRAX models use 12 risk factors which are mentioned above.
Garvan and FRAX models use risk factors to predict fracture risk for
10 years. According to the recommendation of American National
Osteoporosis Foundation (NOF) and International Osteoporosis
Foundation (IOF), individuals at risk of fracture above 20% should
be indicated for treatment. Two models Garvan and FRAX have been
developed and applied in identifying individuals at high risk for
treatment and prevention.
3
In Vietnam, there have been a number of studies on osteoporosis
in specific patient groups, while the study in the population is still
limited. In addition, there have been no studies assessing the risk of
bone fracture in the community, and comparing the prognostic value
of Garvan and FRAX. Therefore, we carried out the topic "Study on
bone mineral density and fracture risk factors in menopausal women
in Rach Gia City, Kien Giang province" with two following
objectives:
+ To survey the BMD with DXA method and determine the rate
of osteoporosis in the community, along with factors related to
osteoporosis in postmenopausal women;
+ To assess the fracture risk in the community through two
Garvan and FRAX models.
The study also compared the prognostic value of two Garvan
and FRAX models and compared with the recommended treatment
indications and current treatment guidelines.
NEW CONTRIBUTION FROM THE THESIS
+ Determine the prevalence of osteoporosis in the community.
An important result and contribution of the thesis is the osteoporosis
scale in the community in Rach Gia City (Kien Giang). The study
indicates that 45% of postmenopausal women have osteoporosis
(osteoporosis, 11.2%) or bone loss (osteopenia, 34%).
+ Determining risk factors related to osteoporosis. The study
found that the following factors are related to osteoporosis: elderly,
age with the first period above 15, weight loss, and infertility.
+ The correlation coefficient of prognostic value between
Garvan and FRAX models is r = 0.7. This result means that the
4
prognostic value of FRAX model explains 49% of the difference in
the prognostic value of the Garvan model. This result shows that two
Garvan and FRAX models have relatively high similarity.
+ Based on the standard of bone fracture risk ≥20%, Garvan
model predicted that 59.2% (122 over 206) had a high risk of
fracture. FRAX model predicted only 7.3% at high risk (15 over
206).
+ In the group of osteoporosis, Garvan model predicted 100%
(23/23) with a high risk of fracture; FRAX models predicted only
60.9% (14/23). In the group with a history of fractures, the Garvan
model has 90% prognosis of high risk (27/30), but the FRAX model
proposes only 30% (9/30). These results show that the Garvan model
is more relevant to clinical reality than the FRAX model.
THESIS OUTLINE
The thesis covers 107 pages, including:
Preamle: 2 pages
Overview: 33 pages
Materials and method:13 pages
Outcome:28 pages
Discussion:28 pages
Conclusion: 2 pages
Recommendation: 1 page
The thesis consists of 42 tables, 6 charts, 9 figures and 129
references (including: 11 references in Vietnamese, 118 references in
English).
CHAPTER 1. OVERVIEW
5
Bone biology
Bone is made up of two main types of tissue: inorganic and
organic. Inorganic ingredients account for 70% while organic
ingredients contribute 22% of bone weight. The main inorganic
component is calcium phosphate hydroxyapatite. Organic ingredients
are mainly collagen type I, accounting for about 85%, and non
collagen proteins (about 15%) such as osteocalcin, osteopontin,
sialoprotein, glycoprotein, proteoglycan, and glaprotein.
Bone is a dynamic tissue created from three main cell groups:
osteoblast, osteoclast, and osteocyte. The two main cells playing an
important role in bone modeling and remodeling are osteoblast and
osteoclast. These two types of cells work together and depend on
each other, not independently. Osteoblast and osteoclast form a
temporary structure called Basic Multicellular Unit (BMU). Each
BMU is about 12 mm long and 0.2 to 0.4 mm wide.
In normal conditions, osteoblasts and osteoclasts smoothly work
together in BMU. When operating normally, the bone mass excreted
is equal to the bone mass produced. However, in postmenopausal
women and older people, osteoclasts are more active than
osteoblasts, which lead to bone loss. Bone loss leads to a decrease in
bone strength and an increased risk of bone fractures. Therefore, on
the biological perspective, osteoporosis can be considered as a
consequence of an imbalance between osteoblasts and osteoclasts.
Osteoporosis
Osteoporosis is a disease whose two main characteristics are reduced bone strength and degraded bone structure, leading to an
increased risk of bone fractures. Bone strength is primarily assessed
6
by bone mineral density (abbreviated as BMD). Low BMD is a risk factor for bone fractures. In 1994, the World Health Organization (WHO) defined Osteoporosis as a disease characterized by reducing
bone mass, damaging the subtle structure of bone, resulting in bone
weakness and consequently increased risk of fractures.
Osteoporosis is therefore diagnosed by measurement of BMD.
As recommended by the World Health Organization, when BMD
drops more than 2.5 standard deviations from the age of 2030 it is
diagnosed with osteoporosis. In Vietnam, there have been several
studies in the past few years on osteoporosis scale in the population.
A community study in HCMC found that the rate of osteoporosis in
women over 60 was about 29%. However, a hospital study in Hanoi
found that nearly 60% of patients were diagnosed.
Bone fractures are a consequence of osteoporosis. Worldwide,
osteoporosis causes fractures every year at least 8.9 million people.
In 2010, 158 million people broke their bones, expected to double in
2040. In Asia, it estimates the risk of neck femoral will increase 2.28
times. Costs for treatment of fracture rise billions of dollars each
year.
Risk factors for osteoporosis
Many clinical epidemiological studies in the world and in
Vietnam have identified a number of risk factors related to
osteoporosis. These factors can be divided into two groups:
modifiable and nonmodifiable risk factors.
Modifiable factors include lifestyle (smoking, excessive
alcohol), poor healthy diet, lack of exercise, reducing sex hormones
(estrogen, testosterone), weight loss, dietary intake of low calcium,
7
vitamin D deficiency, falls, and poor health. Nonmodifiable factors
include elderly, female, hereditary, history of individual fractures,
and family history of fractures.
Risk factor for fracture
All of the risk factors for osteoporosis listed above are also risk
factors for fracture. In addition, the reduction of bone density or
osteoporosis is a considerable risk factor. The above factors affect the
risk of fracture. Therefore, patients who pose more risk factors will
face a greater risk of bone fracture.
Prognosis model
Although BMD is the most important risk factor for bone
fractures, BMD only identifies 55% of women and 25% of men with
broken bones. Therefore, the new trend in osteoporosis is to build
prognostic models for predicting (prognosis) fractures in 10 years
based on each individual's risk factors.
Currently, there are two main prognostic models: FRAX and
Garvan. FRAX model uses 12 risk factors, Garvan model uses 5 risk
factors. Risk factors of FRAX include gender, age, history of bone
fracture, weight, height, femoral neck bone density, family history of
fracture, smoking, alcohol use, corticosteroid use, rheumatoid
arthritis, and secondary osteoporosis. Risk factors in the Garvan
model include gender, age, history of fractures, history of falls, and
femoral neck bone density. However, assessing the correlation
between these two models is still small, and not yet systematic.
The study presented in this thesis is designed to provide
scientific answers to the following questions:
+ How many postmenopausal subjects have osteoporosis in the
8
community in Kien Giang; and which factors are related to
osteoporosis?
+ What is the scale of fracture in the community through FRAX
and Garvan models?
+ Consistency between high fracture risk (through FRAX and
Garvan models) and treatment indication?
STUDY METHOD
The study was conducted in the community of Rach Gia City,
Kien Giang Province. The study duration was from November 2012
to December 2015. The study was designed according to the cross
sectional model.
Study subjects: including 206 menopausal women living in Rach
Gia City, Kien Giang province, agreeing to participate in the study.
Subjects of the study are invited from women's associations and
elderly associations. The subjects were explained about the
objectives and research process and agreed to participate. They were
interviewed at the clinic of Kien Giang General Hospital and Van
Phuoc Clinic (Can Tho).
Measurement of BMD: each subject was measured the bone
density at femoral neck using DXA device labeled Osteocore Station
Mobile (MEDII INK, France) at Van Phuoc clinic (Can Tho). The
value of the BMD was converted to the T index. Based on the scan
results, each woman was classified into one of three groups: normal
(Tscore higher than 1), osteopenia (Tscore in 1 to 2.5), and
osteoporosis (Tscore is equal to or lower than 2.5).
9
Data collection: Each individual provides information related to
anthropology, history of fractures, history of reproduction, lifestyle,
weight, and height. Body mass index (BMI) is calculated from
weight and height and divided into 3 groups: underweight (BMI
<18.5); normal (BMI 18.5 to 24.9), and overweight (BMI 25.0 or
higher).
Estimation of fracture risk. Based on the risk factors for each
subject, the 10year risk of fracture was calculated by FRAX and
Garvan models. As recommended by the World Health Organization
and the US National Osteoporosis Foundation, the probability of a
fracture (10 years) above 20% is considered high risk.
Data analysis: Data were analyzed by descriptive statistics and
logistic regression methods using R. software. Descriptive statistic
methods were used to estimate the prevalence of osteoporosis and
95% confidence intervals. Logistic regression model was used to
assess the relationship between risk factors and history of fractures.
Based on the parameters of the logistic regression model, the odds
ratio and the 95% confidence interval are estimated.
The scale of bone fracture in the community was estimated by
predicting fracture risk with two models FRAX and Garvan.
Percentage of subjects at risk of fracture above 20% is considered as
an estimation of fracture scale in the postmenopausal women
community.
10
STUDY RESULTS
The study was conducted on 206 postmenopausal women. The
median age of the study subjects was 66 (the minimum value was 48
and the maximum was 85 years old). Among 206 subjects, 46%
(n=94) were overweight and 2.4% (n = 5) were underweight.
Results of analysis of Tindex showed that: 113 (55%) had
normal BMD; 70 (34%) osteopenia, and 23 (11.2%) osteoporosis.
The risk factors for osteoporosis are reported in Table 1 below:
Risk factors for osteoporosos
Risk factors OR 95%CI P
Age (+1) 1,15 1,07 – 1,23 <0,0001
BMI (+1) 0,74 0,63 – 0,86 <0,0001
Menstruation after 15 years old 1,35 0,48 – 3,84 0,561
Childless 5,28 1,86 – 14,97 0,003
Menopause before 53 3,42 0,77 – 15,17 0,105
Years since menopause (+1) 1,11 1,05 – 1,17 0,001
History of failing 1,93 0,88 – 4,23 0,102
The family history of fractures 23,9 8,1 – 70,4 <0,0001
Results of logistic regression analysis showed that factors
related to increased risk of osteoporosis were old age, reduced BMI,
infertility, postmenopausal time, and a history of fractures in the
family. For example, every year increased age is related to 15% of
increasing odds of osteoporosis (odds ratio 1.15; 95% confidence
11
interval from 1.07 to 1.23). A history of fractures in the family has
the greatest impact on the risk of fracture.
Among 206 subjects, there were 30 subjects with a history of
fractures. Prevalence of patients with history of fracture was 14.6%
(confidence interval 95% : from 10.4% to 20%). Univariate analysis
showed that the factors related to a history of fracture (statistically
significant) are osteoporosis, infertility, a history of falls, and a
history of fractures in the family.
Table 3.27. The correlation between the history of fractures and
History of fracture
Non fracture before
OR, p
Osteoporosis
Percentage
classification
Quantity
Quantity Percentage %
%
Osteoporosis
10,0
40,0
11
6,25
12
(3,86 – 25,9)
(n= 23)
P < 0,0001
Non osteoporosis
60,0
165
93,7
18
(n=183)
30
Total (n=206)
100,0
176
100,0
osteoporosis
12
Table 3.31. The correlation between history of bone fractures
History of fracture
Non fracture before
OR, p
Status of
Quantit
Percentage
Percentage
childbirth
Quantity
%
%
y
No birth
23,3
14
8,0
7
OR=3,52
(n= 21)
(1,269,8)
Giving birth
p<0,01
76,7
162
92,0
23
(n=185)
Total
100
176
100
30
(n=206)
and no birth
Table 3.34. The correlation between history of fractures and
History of fracture
Non fracture before
OR, p
History of
Percentage
Percentage
falling
Quantity
Quantity
%
%
History of
falling
10
33,3
28
15,9
OR=2,64
(n=38)
(1,106,33)
P=0,03
No falling
before
66,7
148
84,1
20
(n=168)
Total
14,6
176
85,4
30
(n=206)
history of falls
13
Table 3.35. The correlation between fracture and family history
History of fracture
History of nonfracture
History of
OR, p
fractures in
Quantity Percentage % Quantity Percentage %
family
With history of
fracture in
10
33,3
12
6,8
OR=6,83
family (n=22)
(2,4918,7)
No fracture in
P=0,001
family
20
66,7
164
93,2
(n=184)
Total (n=206)
30
100
176
100
of fracture
However, the analysis of multivariate logistic regression shows
that after adjusting for all factors in the model, only osteoporosis is
an independent factor. Accordingly, the subjects with osteoporosis
had an odd ratio of 6.83 (95% in confidence interval from 1.71 to
23.0).
14
Table 3.36. Multivariate regression analysis between history of
fractures and risk factors.
Factors OR 95%CI P
Age ≥ 60 2,85 0,2927,56 0,37
BMI < 18,5 0,27 0,023,47 0,33
1,7122,99 0,007 6,83 Osteoporosis
Menstruation after 15 years old 0,58 0,22 1,49 0,27
Childless 2,36 0,63 6,69 0,15
Menopause before 53 1,00 0,31 3,24 0,99
Postmenopausal time > 10 years 0,46 0,121,74 0,26
History of falls 1,12 0,433,72 0,85
Family history of fractures 2,48 0,8010,74 0,19
Correlation coefficients r=0,70; p<0,01
Figure 3.7. The correlation between prognostic value of fracture
in FRAX model and Garvan model
15
Table 3.37. Predicting the risk of femoral neck fracture by age
group
Frax Model Garvan Model Age group Low risk High risk Low risk High risk (n=206) n (%) n (%) n (%) n (%)
1 16 1 16 < 60 years old
(n=17) (5,9) (94,1) (5,9) (94,1)
2 60 68 126 60 69
(n=128) (1,6) (46,9) (53,1) (98,4)
12 8 53 49 ≥70 years old
(n=61) (19,7) (13,1) (86,9) (80,3)
15 84 122 191 Total (7,3) (40,8) (59,2) (92,7)
Compare with p p<0,05 p<0,05
The results of the prognostic model FRAX and Garvan show
that the correlation coefficient between two models is r = 0.7, and it
was statistically significant (P <0.0001). Using the threshold of
fracture probability> 20%, FRAX model did not detect highrisk
subjects, but Garvan model detected 10.2% (n = 21). Using the
threshold of femur fracture probability> 3%, FRAX model detected
7.3% (n = 15) highrisk subjects, but the Garvan model detected
59.2% (n = 122) at risk high.
16
Table 3.41. Comparison between the indications for treatment of
osteoporosis and high risk based on the prognostic value of
femoral fracture.
Prognostic value Total Osteoporosis p of femoral (n = 206) (n = 23) fracture.
FRAX ≥ 3% 15 (7,3%) 14 (60,9%) p<0,0001
Garvan ≥ 3% 122 (59,2%) 23 (100%) p<0,0001
Table 3.43. Comparison between indicated treatment of history
of fractures and high risk based on the prognostic values of
femoral fractures.
History of Prognostic value Total femoral p of femoral fracture. (n = 206) fracture. (n = 30)
FRAX ≥ 3% 15 (7,3%) 9 (30%) p<0,0001
Garvan ≥ 3% 122 (59,2%) 27 (90%) p<0,0001
In 30 subjects with the history of fractures (having indications
of treatment), FRAX identified 9 subjects (30%) at high risk, but
Garvan identified 27 ones (90%). In 23 subjects with osteoporosis
(having indications of treatment), FRAX model identified 14 subjects
(60.9%) and Garvan identified 23 (100%) at high risk. Therefore, the
Garvan model is more suitable for treatment indications than FRAX
models.
17
DISCUSSION
Osteoporosis and fracture consequences are the public health
burden in the community, especially in menopausal women and
elderly men. Osteoporosis is a "silent" disease that has no specific
symptoms, so identifying highrisk subjects is a difficult fact in
clinical practice. To identify highrisk subjects, understanding the
correlation between risk factors and osteoporosis (and fractures) is
important. Currently, in the specialty of osteoporosis there are two
common models, FRAX and Garvan, which can be used to assess an
individual's fracture risk based on risk factors, and thereby identify
the objects need to be intervened. However, the studies on the
similarity between the two models, and the similarity between
prognostic value and treatment indications in Vietnam are still very
small. This study was conducted on 206 menopausal women which
provided 4 new following information:
+ The rate of osteoporosis and osteopenia in menopausal
women was about 45%;
+ Risk factors related to osteoporosis include elderly, low
BMI, infertility, and a history of fractures in the family;
+ However, when analyzing with the history of fracture, only
osteoporosis (low bone density) was an independent risk factor;
+ Garvan model identified nearly 60% of subjects with high
fracture risk. Garvan model has prognostic value consistent with
clinical treatment indications than the FRAX model.
The new information from this study represents a contribution
to Vietnamese medical literature in the management, treatment, and
prevention of osteoporosis at the community level.
18
Our study focused on postmenopausal women (with an
average age of 66.8). We chose female subjects because women are
more likely to have osteoporosis and fractures than men. In this
group, we found 11% of women are in osteoporosis status (Tscore is
equal to or less than 2.5). The rate of osteoporosis in this study is
somewhat different but is in the average range in comparison to
previous studies. In a study of 504 women in Hanoi, Dang Hong Hoa
et al. (2007) [3] estimated that 9.3% of women with osteoporosis. A
larger study (n = 2232) also on women in Hanoi in 2004 showed that
the rate of osteoporosis was 15.4% [2]. In Ho Chi Minh City, the
study by Ho Pham Shu Lan et al [4] on 970 postmenopausal women
randomly selected in the community showed osteoporosis rate of
29%. In addition, the study of 988 women in Hanoi (in hospital
samples) detected 58.4% of osteoporosis [79]. In summary, the
abovementioned studies indicate that the extent of osteoporosis in
the community can range from 9% to 29%, depending on age and
method of measurement. Our study estimates that the rate of
osteoporosis was 11.2%, which is lower in comparison with the
studies recently reviewed.
The difference in the rate of osteoporosis between studies has
many causes. The studies were based on the sample, and the sample
was selected within the community, so sample fluctuations in
estimating proportions are inevitable. In our study, although the
average rate was 11.2%, the confidence interval was 95% which
ranged from 7% to 16%. It is possible that the subjects in this study
were not highly represented in the community, because they were
recruited from community organizations (Women's Union, Elderly
19
Association), and these people often have better health in the
community, and this reason can also explain the relatively low rate.
The method of measuring bone density may also explain the
differences between studies. We used DXA method (which is
considered the gold standard) with very low technical error. Some
previous studies have also used the DXA method, but may have
different reference values, leading to a situation where the rate of
osteoporosis was very different between studies. In addition, factors
related to lifestyle, anthropology, nutrition, and economic
composition also affect the rate of osteoporosis among the study
samples. It can be said that there is no standard rate for any
population, but all the study results in Vietnam over the past time,
including our research, show that osteoporosis has been a large
public health issue.
We found four factors that are associated with an increased
risk of osteoporosis: elderly, low BMI, infertility, and a family
history of fractures. This result is also quite consistent with many
previous studies in Vietnam and around the world. The higher age,
the greater risk of osteoporosis, and this fact is completely consistent
with the agechanging process of bone density. Bone density decline
after menopause and continues to decline even after the age of 70, so
the frequency of osteoporosis after age 70 is high and completely
consistent with the rule of bone strength decline. The relationship
between age and osteoporosis may also be due to other factors such
as comorbidities and general health conditions that we have not yet
been able to assess and analyze in this study. Therefore, it can be
considered that the increase in age is an indirect indicator that
20
reflects the pathological and health factors rather than a definite
relationship.
The association between BMI and osteoporosis has been
studied a lot in the past, and most studies have shown a positive
association: those with high BMI often have high bone density and
thus reduce the risk of osteoporosis. In this study, we also found that
trend: increased BMI was associated with a lower risk of fracture. In
our study, only 5 subjects had a BMI higher than 30 (obesity), and
most were "overweight". While our study does not allow for a
conclusion about the causality link between BMI and osteoporosis,
this finding means that maintaining moderate weight (BMI between
20 and 27) is probably one of the most realistic solutions for
osteoporosis prevention.
The finding of negative effects of infertility on osteoporosis is
an interesting and clinically significant data. About 10% of subjects
in the study were infertile, and their bone density was lower than the
group having birth about 0.1 g / cm2. This result is also consistent
with many studies worldwide: women with many children have
higher bone density than those with fewer children [95]. However,
the biological mechanism for the association between infertility and
osteoporosis is still unclear. Infertile women often have lower BMI
than women with children, but in this study after adjusting for BMI,
the risk of osteoporosis in women with infertility is still higher than
that of children with children. Another theory is that infertile women
may have lower peak bone density and bone loss than women with
children, so the risk of osteoporosis is increased, but this hypothesis
requires factual data to verify. However, the finding of an association
between infertility and osteoporosis is remarkable, because infertility
21
factors can help clinicians pay more attention to the bone health of
these subjects.
The relationship between the history of fractures in families
and osteoporosis is also an interesting and important finding. Our
findings are consistent with an Australian study showing that women
whose mothers suffered fractures also have reduced bone density
[103]. This is probably due to genetic factors because genetics
explain about 6080% of differences in bone density. Therefore, the
risk of osteoporosis increases in women whose mother or family
members have bone fractures due to bone density decline. This
finding also implies that identifying women at high risk for
osteoporosis needs to pay more attention to their family history.
In this study, we did not track the subjects over time, so it was
not possible to estimate the rate of new fractures, while we could
only assess the history of individual fractures. We observed that 30
subjects (14.6%) had a history of fractures. Although some factors
related to osteoporosis are also associated with a history of fractures,
logistic regression analysis shows that osteoporosis was the only risk
factor associated with a history of fractures. This finding suggests
that osteoporosis is the "closest" factor and may be directly related to
fractures. However, because these subjects had fractures before
measuring bone density, another explanation was that fractures were
responsible for reducing bone density and increasing the risk of
osteoporosis. We cannot determine the cause and effect of the
relationship between osteoporosis and fractures because of the cross
sectional study design.
22
One of the key questions of the study is the burden of bone
fracture in the community. However, since we do not have data on
the incidence of new fractures, we have to apply the prognostic
model Garvan and FRAX to evaluate the fracture scale in the
community. While FRAX model only found 7.3% of women had a
high risk of fracture, Garvan model identified 59.2% of women with
a high risk of fracture (probability of 10year fracture was over 20%).
We believe that the Garvan model's estimation is more suitable for
reality. According to a previous study [50], women aged 60 and older
had a 65% risk of a lifetime fracture. In our study, the average age of
the study subjects was 66, therefore the remaining 60% risk of
lifetime fracture was quite consistent with the fact in the population.
This finding is important because it shows that the fracture scale in
postmenopausal women is quite high and in the context of an aging
population is happening throughout the country, osteoporosis and
fractures will become a burden for the national health system.
An important and interesting finding of this study is the consistency
between the prognostic value of fractures and treatment indications.
In our study, there were 30 subjects with a history of fractures or 23
subjects with osteoporosis, and according to the current guideline,
these subjects have indicated treatment. Among 30 subjects with a
history of fracture who were provided treatment indications, Garvan
model detected 90% (n = 27) at high risk (probability of 10year
femoral fracture is more than 3%), but FRAX model detected only
30% (n = 9) at high risk. Similarly, 23 subjects of osteoporosis
Garvan model detected 100% (n = 23), but FRAX detected 60.9% (n
= 14) high risk. Therefore, we conclude that the Garvan model gives
consistent results with treatment indications over the FRAX model.
23
CONCLUSION
The actual data from the study allows us to come to three
main conclusions as follows:
+ The rate of osteoporosis in postmenopausal women is
11.2% and may range from 7 to 16%. However, the risk of lifetime
fracture in postmenopausal women is 60%.
+ In addition to elderly and low BMI, risk factors related to
osteoporosis include infertility and a history of fractures in the
family. But these factors have an indirect effect on the risk of
fracture. Osteoporosis is an independent factor associated with a
history of fractures.
+ The correlation between FRAX and Garvan models is
relatively good, but only the prognostic value of Garvan model is
highly consistent with the treatment for osteoporosis treatment.
FRAX model estimates that fracture risk is lower than reality, and
therefore cannot be applied in Vietnamese.
24
SUGGESTIONS
From the above conclusions, we suggest:
1. Need a nationalscale study with a large number of sample sizes
and a highly representative sampling method to better assess the scale
of osteoporosis in Vietnam.
2. There should be a national and local osteoporosis prevention
strategy by controlling a number of highrisk factors for fractures,
especially osteoporosis in menopausal women, to reduce the rate of
fractures on these subjects.
3. In the absence of a prognostic fracture model for Vietnam, Garvan
model can be temporarily used to predict the risk of femoral neck
fracture after 10 years in order to warn about the fracture situation.
THE PUBLISHED ARTICLES RELATED TO THE STUDY
Thai Viet Tang, Doan Van De (2018). The study on bone
1.
mineral density and osteoporosis with DEXA method, and risk
factors of osteoporosis in women over 40 years old..
Vietnamese Journal of Medicine, Volume 465, March, (2): 24
27
Thai Viet Tang, Pham Thanh Binh, Doan Van De (2018).
2.
Investigation of fracture rate, fractural risk factor due to
osteoporosis and predicting fracture risk by FRAX and
GARVAN models. Journal of Military Medicine and
Pharmacy, 43(2): 122127.
